Prioritization of Public Transportation Investments: A Guide for Decision-Makers (2021)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

16 Building Successful Practice 3.1 Attributes of Successful Prioritization Practice National research, guidance, and state of the practice reviews pro- vide several lessons regarding attributes of successful practices in investment decision-making (Figure 7). Much of the existing guidance on multi-objective investment prioritization is also applicable to public transportation. However, some considerations are unique to transit that merit special consideration. Dening Objectives and Measuring Progress It is important to use care in dening an organization’s fundamental objectives when setting up an investment prioritization approach. at is, what outcome is the transportation agency trying to achieve through its investments? e set of objectives should be comprehen- sive, and the individual objectives should be distinct from one another (non-overlapping). is issue has been explored extensively in multi-objective decision-making and is detailed in resources such as Keeney and Raia (1993). With objectives dened, performance measures are used to quantify progress toward meeting objectives. ere are many challenges in dening appropriate performance measures, particu- larly when attempting to prioritize across modes, as discussed in the following sections. Ideally, prioritization relies on quantitative measures that reect the progress toward a given objective. For instance, a reduction in hours of travel delay can be used to help quantify the objective of improving mobility. In practice, however, it can be dicult to dene and quantify such measures, especially for prospective future investments for which detailed data are specula- tive or unavailable. Guidance on the selection of dierent types of criteria, whether quantitative or qualitative, is summarized in Section 3.4. Creating Structure for Comparing Across Projects and Assessing Multiple Areas of Performance Eective prioritization requires structure, repeatability, and accountability to objectives and stakeholders. To achieve this, it is important to include stakeholders in the process and tailor the process to address stakeholder needs. Advisory groups play an important role in many invest- ment prioritization processes and should include representatives from subsidiary government agencies or agency units that will be governed by the prioritization process. For statewide priori- tization, this may mean representation from DOT district oces, modal divisions, and MPOs C H A P T E R 3 Successful transit prioritization incorporates a wide range of objectives and measures that reflect transit benefits: travel efficiency (travel-time cost savings, congestion reduction, and reliability), affordable access, enhanced economic development, safety, and reducing negative impacts of travel on the environment.

Building Successful Practice 17   or regional planning organizations (RPOs). Regional prioritization by an MPO typically includes representation from local member govern- ments and public transit agencies. In both cases, community leader- ship representing specic perspectives or areas of expertise are oen included as well. In addition, an investment prioritization process must dene not only objectives and measures but also an approach for comparing across projects and for incorporating information on multiple areas of performance. One approach to structuring an investment prioritiza- tion approach is to rely on a benet-cost analysis (BCA) framework and monetize measures where possible, using clearly documented and justi- ed valuation factors. is approach helps align measures with project value. BCA frameworks also make use of the existing body of knowl- edge to address trade-os between dierent objectives (e.g., comparing investments to improve safety with other investments to improve mobility). However, even when a BCA and/or monetized measures are used, most agencies responsible for transit prioritization also consider objectives that cannot be easily quantied. us, many of the examples in the literature include a combination of highly quantitative and more qualitative approaches. Both quantitative and qualitative measures can be combined within a multicriteria scoring framework where dierent categories of outcomes are assigned weights that are then used to aggregate results and compare across projects. Considerations for Transit Investment Prioritization While investment prioritization best practices are broadly applicable across modes, the following areas merit consideration when designing a prioritization process that will apply to public transportation: • “Widening the lens” to incorporate a wide range of objectives and measures to ensure that public transit benets are suciently captured. Criteria that are particularly eective in this area are detailed in Section 3.2. • Considering the data and analytical capabilities necessary to support eective measurement of transit benets at varying levels of complexity, as detailed in Sections 3.3 and 3.4. • Keeping a close eye on issues related to multimodal comparability and modal bias. Tactics to avoid modal bias are detailed in Section 3.5. Carefully define investment objectives Measure progress toward objectives Leverage benefit-cost framework to compare among objectives, scale relative to cost Integrate qualitative approaches for objectives that are important but not easily quantified Context matters – consider the decisions that are to be informed, types of investments, time frame, available data Work toward structure, repeatability, and accountability to objectives and stakeholders Source: EBP, based on a review of literature from research and practice. Figure 7. Attributes of successful investment prioritization practice. Benet-Cost Analysis According to the U.S. DOT, benefit- cost analysis (BCA) is a “systematic process for identifying, quantifying, and comparing expected benefits and costs of a potential infrastructure project” (U.S. DOT 2021). One advantage of BCA is that it uses quantitative and/or monetizable metrics to effectively address trade-offs between different objectives. However, most agencies also consider objectives that cannot be easily quantified or included in BCA.

18 Prioritization of Public Transportation Investments: A Guide for Decision-Makers 3.2 Decision Criteria That Capture the Benets of Public Transportation e selection of decision criteria that fully capture the benets of public transportation helps ensure that funds are directed toward investments that will be successful and sustainable in the long run. To that end, Table 3 and Table 4 oer a summary of criteria that have posi- tively aected the competitiveness of public transportation projects by capturing their intended and likely outcomes and benets. ese were identied through interviews with practitioners. e criteria include both quantitative and qualitative scoring processes developed for a variety of contexts, including investment prioritization decisions internal to a public transit agency or state, regional long-range plans involving multi-agency coordination, and shorter-term priori- tization for state or MPO transportation improvement programs. In each case, these decision criteria are guided by the prioritizing agency’s goals and plans; applicable legislative, regulatory, or policy requirements (including federal performance management guidelines); and public/ stakeholder input. e criteria in Table 3 are typically only applicable to public transportation investments, while the criteria in Table 4 represent cases where transportation agencies with multimodal jurisdiction can use the same criteria in evaluating more than one mode. Criteria repeatedly cited in interviews as being successful are indicated with a star (★). Note that decision-makers use evaluation criteria to both (a) capture public transportation benets for users and society and (b) ensure public transit investments are successful and sustainable in the long run. Examples in the latter category include measures such as service productivity, SGR, market potential (where full ridership forecasting/travel analysis is not conducted), and land-use compatibility. Decision criteria help direct funds toward investments that will be successful and sustainable. A wide range of decision criteria can positively affect the competitiveness of public transportation projects by capturing their intended and likely outcomes and benefits. Criteria Type Description Service coverage/ Completeness Includes measures that capture and emphasize the comprehensiveness of public transportation services geographically or temporally, such as route miles or total system hours. Eligibility for value capture financing Captures public transportation s unique ability to generate real estate value by increasing accessibility, which can also generate funding opportunities through a mechanism like tax increment financing. Integration of transit services Measures a project s ability to integrate processes or services across public transit agencies in a state or region. Productivity Highlights the potential for a project to increase operating efficiencies by, for example, increasing vehicle occupancy or reducing costs per passenger mile. The American Public Transportation Association (APTA) defines productivity as “the ratio of units of transportation output to units of input (consumed resource); for example, vehicle miles (vehicle kilometers) per operator hour, or passenger miles (passenger kilometers) per unit cost of operation” (APTA 2019a). Quality of service Can include measures that capture how crowded public transit vehicles are as well as reliability measures like on-time performance (transit only) or level of travel-time reliability and planning-time index (all modes). Table 3. Transit-specic decision criteria that capture the benets of public transportation.

Building Successful Practice 19   Criteria Type Description «Accessibility Measures such as access to jobs for disadvantaged populations or access for zero-car households capture the importance of public transportation for low-income individuals and for accessibility more broadly. «Congestion/ Mobility Highlights public transit’s ability to relieve congestion or help people avoid congestion through measures related to reducing person-hours of delay, reducing travel time between major activity centers, or increasing person throughput, often at the corridor level. «Cost- effectiveness/ System preservation Performance measures that emphasize maintaining existing assets as a long-term cost savings mechanism tend to make transit more competitive. These measures could include a comparison of costs avoided to annualized costs of a project, breakeven year estimates, etc. For expansion projects, this approach means ensuring that costs incorporate full life cycle costs, not just up-front capital expenditures. «Economic impacts Supports an understanding of transit’s role in the economy through measures related to supporting economic development, such as economic productivity, output, or return on investment. «Environmental quality Measures related to avoiding or reducing impacts to natural and cultural resources may capture positive externalities of transit use (overlaps with public health and quality of life). These measures may address both localized environmental impacts and broader climate change-related goals of reducing greenhouse gas emissions. «Land-use compatibility Favors projects that have increased potential for success because of compatible land use (existing or planned) in the vicinity of a proposed project and/or projects that support efficient land development patterns. Public health and quality of life Measures related to air quality, livability, and support for active transportation/healthy lifestyles (including walking to transit) reflect the advantages of transit capital projects. «Regional integration and coordination Measures a project’s cohesiveness with broader regional plans from partner agencies; reflects the interagency nature of some transit planning. «Social equity/ Environmental justice Expansion of public transportation services has the potential to compensate for historic underinvestment in specific communities and to provide options to the mobility disadvantaged (whether due to income, location, race, ability, or the intersection of these). Viability/feasibility Evaluate whether the amount allocated can fund a viable project or set of projects, thus maximizing the cost- effectiveness of transit capital investments. Table 4. Multimodal decision criteria that capture the benets of public transportation. 3.3 Data and Capability Requirements To support the application of investment prioritization criteria, transit agencies, MPOs, and state DOTs rely on a range of tools, data sources, capabilities, and processes to evaluate public transportation projects. e following section provides an overview of common evalu- ation methods and data sources employed in transit prioritization, along with a discussion of emerging data and tools. is information provides a starting point for decision-makers to assess their data and analyt- ical capabilities to support the criteria discussed in Section 3.2. Keep in mind that there are always trade-os between seeking the best data possible to characterize a project’s expected

20 Prioritization of Public Transportation Investments: A Guide for Decision-Makers benets and the time and eort involved in acquiring or developing such data. For this reason, Section 3.4 describes ways in which evalua- tion criteria can be implemented to various degrees of quantication, including relying on structured qualitative information where data, methods, or tools are either unavailable or insucient. Additionally, as outlined in Section 4.1, it can also be helpful to investigate which agency is best equipped to provide a given type of data and how data sharing may support prioritization. For example, transit asset informa- tion is typically managed by transit agencies, whereas demand models and forecasts are more oen the purview of MPOs and in some cases, state DOTs. Evaluation Methods Traditional analytical approaches to evaluating transportation invest- ment include needs models, impact models, BCAs, and multicriteria analysis (Duncan and Schroeckenthaler 2017): • Needs models predict the impact of programmatic funding levels and compare needs against the economic cost to users and funding agencies. Most needs models use BCAs, including life cycle asset costs. Accurate representation of life cycle costs supports the reallocation of resources among spending programs (e.g., preservation, modernization, and capacity). • Impact models assess the economic and societal impacts of allocation scenarios or projects over time. • BCA monetizes all the benets and costs of a project or program. Some societal values, like equity, are dicult to incorporate into such models. Although a benet-cost ratio is a common performance measure, transit benets can be more dicult to monetize for use in BCA. • Multicriteria analysis is a more exible model that can incorporate qualitative, quantitative, and monetary values into its analysis. Some transportation agencies use decision-making soware in their eorts to set weights for multicriteria analysis. Data Sources All evaluation methods and performance criteria described above rely on data sources to support investment evaluation and prioritization. Specic data sources include: • Project costs: Up-front capital and long-term operating and maintenance costs are necessary for analyzing the cost-eectiveness of proposed projects by comparing measures of desired project outcomes to costs. Cost data may be developed through a combination of engineering- based cost estimation procedures and a review of historical expenditures. • Asset condition: Asset condition data for public transportation vehicles and facilities (e.g., age, mileage, expected service life) are necessary for determining eligibility for SGR funding and/or service impact/asset condition scores in investment prioritization schemes. Asset management systems oen house this information. Insucient asset information and analyt- ical capabilities can be limitations in this area. • Plan goals: Goals and principles from planning documents are a form of information (data) that supports qualitative analysis of public transportation projects and their alignment with community objectives. • Ridership and transportation performance forecasts: Forecasts are necessary for any measure that considers user impacts of investments, rather than just current or baseline future conditions (i.e., without the proposed improvement). However, establishing consis- tent approaches to estimating public transportation ridership can be a challenge, especially Transit Economic Requirements Model (TERM) The FTA uses a model called the Transit Economic Requirements Model (TERM) to assess nationwide needs for future service. FTA uses TERM to make condition and performance reports to Congress. FTA has also released a version of the tool—TERM Lite—that can be used by individual transit agencies to support transit asset management and state-of- good-repair analysis (FTA n.d.).

Building Successful Practice 21   when considering potential new riders. e next section discusses travel demand models. In some cases, simplied elasticity-based approaches are also used to forecast the results of changes to service or cost. • Spatial/demographic data: ese data—for example, census data on concentrations of low-income, minority, or ADA populations—are essential for many measures of social equity or environmental justice. Spatial/demographic data are also needed for measures of land-use compatibility, public transportation market potential, physical impacts (i.e., right-of-way), some assessments of economic develop- ment opportunities (i.e., access to commercial districts), impacts to natural habitat (i.e., wetlands, oodplains, etc.), and the built envi- ronment (i.e., areas identied on the National Historic Register). Analysts oen rely on spatial analysis methods within geographic information system (GIS) soware to process this type of data. • System data: System data support measures, like public transporta- tion service hours, service population, corridors served, and vehicle utilization, are generally easier to compile and more reliable than trip data. Depending on the public transit agency size and reporter type, existing protocol for reporting to the National Transit Database (NTD) can support this type of measure. • Valuation factors: BCA specically requires valuation factors such as the value of time, avoided emissions, or crashes. Many agencies rely on guidance and parameters published by U.S. DOT for this purpose (U.S. DOT 2021), but some also use their localized parameters. Travel Demand Models and Beyond – Opportunities from New Data and Tools Across dierent evaluation methods, travel demand models are oen used to estimate future transportation demand and system performance based on the supply of services and networks and demographic/socioeconomic conditions. is can include custom-built regional or state- wide multimodal models or simplied models such as the FTA Simplied Trips-on-Project Soware (STOPS) (FTA 2020). Most MPOs have or use some form of travel model. State DOTs and public transit agencies, on the other hand, oen do not have ready access to travel demand models and therefore may instead rely on modeling conducted by MPOs or on alternative sources of data and forecasts when evaluating projects. Traditional travel demand models are “4-step” models that follow four sequential steps: trip generation, trip distribution, mode split, and trip assignment. Newer activity-based demand models are adopted by researchers and larger MPOs (and in more limited cases, DOTs) who have more technical capacity and can support complex analysis related to trip purpose. e degree to which these models accurately represent public transportation activity can vary signicantly. Despite their strengths, some travel demand models do not adequately forecast ridership and mode shi responses to new or enhanced transit services. is challenge is most salient in regions where the existing transit market is small and where the characteristics and behaviors of existing transit users may not be particularly representative of the characteris- tics and behavior of potential new transit users that could be attracted National Spatial and Demographic Data The U.S. Census Bureau provides national data resources that can be helpful in characterizing demographics, business activity, and travel patterns, including the American Community Survey (U.S. Census Bureau n.d.-a) and the Longitudinal Employer-Household Dynamics data sets (U.S. Census Bureau n.d.-b) Simplied Trips-on-Project Software (STOPS) FTA has developed a simplified method to quantify the measures used by FTA to evaluate and rate New Starts and Small Starts projects (FTA 2020). STOPS is a limited implementation of the conventional “4-step” travel model. STOPS replaces the standard trip generation and trip distribution steps with the census tabulations to describe overall travel markets. It also replaces the traditional coded transit network with standard transit-services data in the General Transit Feed Specification (GTFS) format.

22 Prioritization of Public Transportation Investments: A Guide for Decision-Makers by meaningfully improved or expanded service. While travel demand, model-based methods are core to many planning and evaluation processes, particularly for MPOs, these limitations should be considered when developing transit investment prioritization methods. Data from bus and rail intelligent transportation systems (ITS) are another increasingly adopted resource for public transit service planning and management. ese include: • Automatic vehicle location (AVL): AVL systems are vehicle-tracking systems that identify and record the location of transit vehicles over time. Real-time transit vehicle movements provide detailed information on travel patterns, speeds, congestion, and reliability. • Automated fare collection (AFC): AFC systems consist of fareboxes that are installed in vehicles and at station fare gates that interact with fare cards (including smart cards) issued to passengers to collect fares (Sanchez-Martinez and Munizaga 2016). • Automatic passenger count (APC): APC systems consist of a variety of technologies that count passengers in the transit system. APC and AFC systems are both helpful for metrics related to trips generated, transit quality, or trips relieved (i.e., standing room only trips that become seated trips). • General Transit Feed Specication: GTFS data consist of transit system and schedule data that can provide insight into transit network coverage and are frequently used to estimate travel times in accessibility analysis (Hemily 2015). GTFS data can also be used to support soware tools that evaluate level of service by population or identify locations with clustered transit stops suitable for a cost-eective investment. • Smart location data: Location records from smartphones and other devices and location- based services are starting to be used to construct detailed mobility traces showing travel across modes from origin to destination and may be used more in the future. ese data sources enable detailed analysis of origins and destinations of transit users that can be used to estimate the number of existing transit users who are impacted by transit improve- ments and can provide insight into which population segments are expected to be impacted (Sanchez-Martinez and Munizaga 2016). However, analyzing these data sets can require signi- cant time and sta capacity, as well as an understanding of data coverage and quality limitations. Since smart cards and AFC systems are relatively new and administered dierently across all transit agencies, there are limitations on available data and processing methods. Shared, stan- dardized tools for analyzing these data sets could aid in their application to project prioritiza- tion. e debate over ownership and privacy constraints, in particular for passenger-related data, can also limit the use and process of transit data. ere are other emerging methods and technology-enabled data sources that are increasingly of interest because of their potential to provide more comprehensive and accurate information for transit evaluation. ese include the use of network analysis outside of a travel demand model to explore changes in accessibility with and without proposed projects. e outputs of land-use models can support the development of land-use scenarios and promote understanding of land-use impacts of public transportation investments. However, land-use modeling can be very computationally intensive. 3.4 Evaluation Criteria Come in Many Forms For all categories of benets or evaluation criteria, there are dierent implementation options that have varying levels of complexity. Choosing an approach is not an “all-or-nothing” decision. In prac- tice, it is helpful to mix dierent forms of evaluation criteria based It is better to account for an important objective in a simplified manner than to not account for it at all.

Building Successful Practice 23   on available information, tools, sta capacity for analysis, and whether a particular objec- tive lends itself to quantication. Options in this area and their pros and cons are shown in Table 5. Similarly, project evaluation and prioritization can be based on: • Condition information (current or forecast future): An example of condition information would be the measure of current on-time performance or present or future land-use mix. • Project attributes: For example, a project might be given a certain number of points within an evaluation framework if it includes transit priority measures such as dedicated lanes or queue jumps. • Project impact: Estimation of project impacts, such as ridership increases, travel time savings, or improved accessibility based on the characteristics of a project and how it interacts with the rest of the transportation system, provide more detail on expected outcomes than typically available otherwise. Nevertheless, impact assessments are the most resource-intensive to develop and depending on available data and tools, may or may not be feasible or desirable. Option Description Pros & Cons Qualitative input Decision-factor considered through qualitative or descriptive analysis Pros: • Does not require data collection or processing • Addresses hard-to-quantify objectives • Can be used to integrate expert knowledge Cons: • Subjective and hard to replicate consistently • Relationship to decision outcomes may not be clear Ordinal scoring Scoring of alignment with criteria along a point-based scale Note: 3-point scales (Low/Medium/High) provide minimum resolution, while more points (e.g., 5- or 7- point scales) provide a more meaningful resolution Pros: • Simpler than full quantitative evaluations • Can integrate formalized guidelines for how to apply ordinal scores, which introduces greater objectivity and reproducibility • Helpful in data-poor environments or for hard-to-quantify outcomes Cons: • Can still be subjective • Requires great care in definition and application of scoring rubrics Quantitative measures Measures that represent the magnitude of alignment with objectives (e.g., travel-time savings, monetized benefits) Pros: • Increased objectivity, replicability • Can address the full spectrum of potential relative differences across projects, allowing for more comparability Cons: • May be constrained by data, analytical capacity, or accuracy • Not all objectives can be easily quantified • Can be resource- and time-intensive Table 5. Options for implementation of evaluation criteria.

24 Prioritization of Public Transportation Investments: A Guide for Decision-Makers It is usually better to account for an important objective in a simplied manner than not to account for it at all. 3.5 Avoiding Modal Bias While Section 3.2 enumerates decision criteria that capture the benets of public transporta- tion, it is equally important for transportation planners and analysts to understand factors that may unfairly or inherently preference other non-transit modes and therefore negatively aect the competitiveness of public transportation projects in the investment prioritization process. Decision-makers should consider strategies to broaden these factors to make them more appli- cable to public transit projects or to pair them with criteria from Table 3 and Table 4 to ensure full coverage of transit-relevant goals. Table 6 summarizes measures that practitioners identi- ed in interviews as potentially introducing modal bias into public transportation investment prioritization, as well as strategies for overcoming each challenge. While not a comprehensive list, it is indicative of potential sources of bias of which those who design investment prioritiza- tion processes should be aware. Potential sources of modal bias against transit in investment prioritization include: • Greater difficulty in quantifying or monetizing relevant benefits • Mismatch of cost perspectives (up-front versus ongoing operations) • Inclusion of criteria that do not apply to transit • Insufficiently capturing cross-modal benefits Description/ Public T ransit Challenge Example Strategy for O vercoming Challenge B enefit- Cost R atio: Monetiz ing benefits is often more complex for public transportation than for other modes, particularly highway and safety projects. To overcome this challenge, San Francisco’ s Metropolitan Transportation Commission (MTC) includes a broad set of multimodal factors within its BCA. In addition, it does not label any project, whether transit or non- transit, as “ low performing” regardless of the BCA results, in recognition of the fact that some projects with a low BCA can be enhanced with further supportive policies. The evaluation results enable MTC to initiate a conversation with project sponsors (localities, public transit agencies) to try to improve project performance. B enefit- Cost R atio: BCA methods fail to incorporate all possible benefits. MetroPlan (Flagstaff, Ariz ona, MPO) recogniz es the inability to evaluate and monetiz e the full range of benefits of projects, including public transit. BCA is instead used as a screening criterion to determine which projects advance to a more in-depth assessment of congestion relief, arterial density, mode choice, safety, and economic development. B enefit- Cost R atio: Benefit-cost ratios do not readily communicate the difference in project scales. The Delaware Valley Regional Planning Commission (DVRPC) addresses this challenge by considering not only total benefit-cost ratios but also benefit-cost per person ratios as well as total benefit measures. Similarly, the ATL presented evaluation results in groupings of high impact/high cost, high impact/low cost, and low impact/low cost. Cost Efficiency: Public investment costs for transit reflect both up-front and ongoing service operation costs, whereas costs for roadways do not reflect vehicle-associated costs, such as car ownership and operations. Methods that account for user costs savings for both public transit and other modal users, including ideally both marginal and fixed vehicle operating costs, can address this issue. Benefit costs methodologies do typically account for this, at least in part. Table 6. Strategies for overcoming challenges with measures and criteria that are unfavorable to or problematic for transit project evaluation.

Building Successful Practice 25   Freight: Freight impacts may be included in multimodal prioritization but are generally not applicable to public “transit” projects. The intent of including freight-related metrics is generally to address the importance of freight to the economy. Including broader economic development measures alongside freight-related measures can capture the impact of transit on the economy. In this way, both highway and transit projects can be considered from a more equivalent economic development perspective. Local Match: Transit practitioners report that differences in local match expectations for federally funded highways versus transit major capital projects can put public transit at a competitive disadvantage. A broader notion of transit “fundability” can account for a wider range of funding sources contributing to project viability/feasibility, such as value capture. Description/Public Transit Challenge Example Strategy for Overcoming Challenge Safety: Planning-level crash modification factors often assign large benefits to highway projects because of the large impact of highway crashes, while public transit fatalities are much less common to begin with. Inclusion of injuries alongside fatalities can remedy this potential bias; periodic measurement of public transit customer sentiment of perceived safety and security can also expand the lens with which safety is considered; finally, consideration of reduced crash exposure from mode shift (e.g., from driving to transit) can capture public transportation benefits more completely. Transit Supportive Infrastructure: The prioritization of non-transit projects may not adequately consider the potential benefit of integrating public transit considerations into other modes. An effective practice for promoting public transit is to recognize the transit benefits of non-transit projects, even for highway projects. While a given project may focus on improving a road used primarily for road traffic, it can still support use by other modes through features such as transit signal prioritization. Table 6. (Continued).